Epidemiology of A influenza viruses - Insights in the molecular epidemiology and antigenic cha

Influenza A viruses are present worldwide and can infect a wide variety of birds, particularly aquatic migratory species (such mallards) and domestic birds (quail, chicken, turkey, etc), aquatic mammals (whales, seals), terrestrial mammals (pigs,

Chapter 1

horses, dogs, cats, etc.) and humans (Webster et al., 1992). The lineages of these viruses are strongly related with the infected host being possible in many instances to establish philogenetically whether or not a given isolate circulates primarily in a given species (e.g. human viruses, avian viruses, etc.). Therefore, the epidemiology of influenza A viruses depends considerably on the ecology of its hosts.

Historically, it has been hypothesized that the restriction of host range for influenza A viruses depends firstly on the type of cell receptors, α-2,3 or α-2,6 predominant in a given species and, secondly, on the affinity of the HA for one type of receptor or the other (Rogers and Paulson, 1983). In the case of avian species and the horse, the predominant receptor for influenza A is the α-2,3, while in the case of other mammals, is the α-2,6 linked receptor. Pigs have both α-2,3 and α-2,6 receptors in the respiratory tract. Until recently, it was assumed that avian influenza viruses necessarily needed adaptation in pigs to become transmissible to humans (Kida et al., 1994). Nowadays, it is know that humans possess both type of receptors in sufficient numbers to grant that at least some avian influenza A strains can be transmitted directly from birds to humans (Reviewed by Imai and Kawaoka, 2012). Actually, direct interspecies transmission of influenza A viruses have been reported from birds to humans and pigs. The episodes of avian H5N1 (Subbarao et al., 2000) are a practical demonstration of this. Also, human viruses easily circulate in pigs and swine viruses can infect humans (Pensaert et al., 1981; Subbarao, 2000; Fouchier et al., 2004; Adiego Sancho et al., 2008; Howden et al., 2009; Smith et al., 2009b)

Aquatic birds are the central elements in the epidemiology of influenza since they can be infected by any subtype of influenza A (Olsen et al., 2006) acting thus as the main

Chapter 1

reservoirs of influenza A in Nature. In contrast, only a few virus subtypes are able to establish in mammals (Figure 1 and 4). For example, for horses only two subtypes have been detected so far: H7N7 and H3N8 and, of these, H7N7 is thought to be extinct (Webster et al., 1993; Bryant et al., 2006). Marine mammals are mostly infected by avian viruses of different subtypes, for example: H3N3, H4N5, H7N7, H13N2, H13N9 (Webster et al., 1981; Hinshaw et al., 1986; Callan et al., 1995; Anthony et al., 2012).

In the case of dogs, infections by H3N8 are the commonest with other human and avian viruses sporadically reported (Gibbs and Anderson, 2010; Rivailler et al., 2010;

Damiani et al., 2012; Lee et al. 2012; Park et al., 2012).

Figure 4. Influenza A virus reservoirs and transmission range. Aquatic birds are the reservoirs of influenza A in nature. Transmission from aquatic birds to other species has been reported. Also, direct transmission between human and pigs, poultry to human, domestic poultry and horses to dogs, pigs to minks and human to dog and minks can be found described in the literature.

Chapter 1

In human and swine only some subtypes have been detected, mainly H1 and H3, but a wide range of variants and reassortants can be found infecting human and pigs.

Characteristics of avian, human and swine influenza A viruses are reviewed in detail in the following sections.

1.4.1. Influenza in avian species

As commented before, birds can be infected by all influenza A subtypes. However, it is in the Order Anseriformes (duck, geese, swan, etc.) and in the Order Charadriiformes (gulls, terns, waders, etc.) where the widest variety of influenza A subtypes have been detected (World Health Organization, 1980). Both orders constitute the major reservoirs of influenza A (Olsen et al., 2006). Subtypes H13 and H16 seem to infect more specifically some Charadriiformes and it has been suggested that these subtypes belong to a genetically isolated branch of avian influenza A viruses (Fouchier et al., 2005;

Olsen et al., 2006).

The spread of influenza in birds is related with the routes of the migratory species of the abovementioned orders (Olsen et al., 2006). In summer and early fall the prevalences of influenza A in migratory birds is higher, probably because the season of births takes place in summer, and thus the bird population receives a flow of new susceptible animals during those months. In contrast, in spring the prevalence of influenza-infected birds is lower because the population comprises older animals that probably have been infected before and were thus immune. As a result, northwards migration in spring contribute little to the spread of influenza while southwards migration in autumm is a serious source of new influenza viruses (Olsen et al., 2006).

Chapter 1

As a result of long-term isolation of hosts depending on the migratory flyways avian influenza A viruses have evolved in two main lineages, the Eurasian and the American (Donis et al., 1989). Nevertheless, viruses carrying genes from both lineages have been detected (Liu et al., 2004; Wallensten et al., 2005; Koehler et al., 2008), indicating that this separation is partial and that the epidemiology of influenza A viruses of birds is probably more complex than thought.

1.4.2. Influenza in humans

The human influenza has by two main epidemiological presentations: seasonal epidemics and global pandemics. The seasonal form occurs when a human influenza virus (HuIV) circulates endemically. In that case, the pre-existing immunity of the population selects viral variants that harbor antigenic changes –because of the antigenic drift- allowing them to escape from the immune system. The time needed for the rise of a new variant of an already circulating strain in a non-naïve population is of about one year. Thus, after that period a seasonal epidemic will take place (White and Fenner, 1994). That is one of the reasons for the need of constant actualization of HuIV vaccines.

Pandemic influenza is a phenomenon representing the global spread of a new influenza A strain. that requires three conditions: 1) The generation of a new strain against which the population do not have any pre-existing immunity; 2) the adaptation of the new strain to replicate efficiently in the human host and, 3) the strain has to be easily transmitted between the hosts (World Health Organization, 2005).

Chapter 1

Usually, the spread of the new epidemic strain is very fast and in about six months the virus can be detected worldwide. This has been observed in all pandemics since Spanish flu of 1918 (Cox and Subbarao, 2000; Taugenberger and Morens, 2006). After this first phase of global spread, secondary waves of spread will occur. These secondary waves are influenced by the previous development of immunity in the human population and the subsequent outbreaks will affect smaller numbers of people. At this point, the pandemic strain has usually displaced seasonal strains and later on becomes a seasonal virus in a new inter-pandemic phase.

At present, the three main subtypes of influenza A viruses circulating in humans are the H1N1, the H2N2 and the H3N2 (Morens et al., 2010). In the last 100 years. Two H1N1 strains that have been involved in pandemics; the oldest is 1918 Spanish flu. It has been suggested that that virus was directly adapted from birds to humans (Smith et al., 2009a). The second H1N1 strain is the 2009 virus originated in North America, which has been demonstrated to be closely related with triple reassortant H1N1 (trH1N1) (see section 1.4.3.7) swine viruses circulating in North American farms and that became adapted to human (Smith et al., 2009b). Russian flu of 1977 was directly related with the seasonal H1N1 circulating in the 50’s and derived from the 1918 strain.

In the 1957 a new pandemic occurred. In that case, was a H2N2 virus reassortant containing the HA, NA and PB1 genes from H2N2 avian viruses and the PB2, PA, NP, M and NS from the previously circulating H1N1. In 1968, the H3N2 virus causing the so-called Asian Flu pandemic was the product of a new reassortment between the H2N2 of 1957 an avian H3. The new virus acquired also the PB1 gene segment from this avian

Chapter 1

virus. It is unclear in what species those reassortments took place although it has been hypothesized that pigs could have played a role in the generation of those pandemic strains (Ito et al., 1998; Suzuki et al., 2000).

Human infections with avian H5, H7 and H9 viruses directly transmitted from birds have been reported (Yuen et al., 1998; Saito et al., 2001; Fouchier et al., 2004). In most cases, those infections have been mild except for the H5N1 strain spreading from Hong Kong area since 1997 which fatality rates for humans were very high (Yuen et al., 1998).

1.4.3. Swine influenza

1.4.3.1. Diversity of swine influenza viruses

For a proper understanding of the epidemiology of swine influenza it is important to understand firstly the ecology of domestic pigs. In many countries, pig production is concentrated in commercial units managed under industrial criteria. Thus, pigs are often confined indoors in large groups and exportation of live animals between regions or even countries (e.g. Netherlands to Spain; Denmark to Hungary) take place between those commercial units where pigs are produced but live pigs rarely travel from continent to another. Nevertheless, familiar pig production still exists in substantial numbers in non-EU Eastern Europe countries, Asia, Latin America and Africa. In industrial units, new susceptible animals are continuously introduced because of the short life span of fattening pigs and the high replacement rates (30%-50%) of sows.

These two factors explain at least partially why only three main subtypes of swine

Chapter 1

influenza viruses (H1N1, H1N2 and H3N2) are found and also may explain the divergent evolution of swine influenza viruses in two different continents. Moreover, in spite that mutation rates are similar in human and swine influenza viruses, the antigenic drift is slower for pig viruses, probably because of the constant flow of the new naive animals, a fact that decreases the selective pressure created by antibodies against circulating subtypes (Noble et al., 1993; De Jong; et al., 2007).

As long as new sequences of SIV are studied, genetic diversity of influenza viruses is found in a given geographical region (Vincent et al., 2009; Kuntz-Simon and Madec, 2009; Moreno et al., 2012; Vijaykrishna et al., 2011). Thus, phylogenetic trees based on nucleotídic composition of the SIV show that, if compared with human influenza A viruses, in pigs there are more differentiated evolutionary lines but with less genetic drift regarding common ancestors. This genetic heterogenicity observed in SIV has been also reated with antigenic heterogenicity in H1N1 strains (de Jong et al., 2001).

As mentioned above, antigenic shift plays an important role in the generation of new SIV strains. In fact, most of current SIV strains are products of reassortment events (Olsen et al., 2002; Kuntz-Simon and Madec, 2009) and it has been demonstrated that swine influenza viruses circulating in Europe present a high reassortment rate (Lycett et al., 2012).In the next sections the distribution of SIV in the different continents will be reviewed.

Chapter 1

20 1.4.3.2. SIV in Europe

Genetic and antigenic diversity of SIV in Europe (Figure 5) have distinctive features.

Earliest SIV isolations in Europe were done between 1938 and 1940 (Lamont, 1938;

Blakemore and Gledhill, 1941). Those isolates were very close to the predominant 1918 Spanish flu-derived H1N1 and could be differentiated from the American H1N1 SIV, a fact that suggested different evolutionary lines (Neumeier et al., 1994). That H1N1 remained predominant in European pigs until 1976, when the North American classical swine H1N1 (csH1N1, see section 1.4.3.4) was introduced in Southern Italy and spread to other countries of Europe (Nardelli et al., 1978; Masurel et al., 1983; Abusugra et al., 1987; Roberts et al., 1987). There are evidences that the 1977 Russian flu virus also spread in the European swine as seen in serological studies performed in several countries (Yus et al., 1992; Brown et al., 1993b). In 1979, an H1N1 strain of avian origin entered in swine of Belgium and Germany (Pensaert et al., 1981), and spread across Europe. The new H1N1 strain displaced the previous csH1N1 (Schultz et al., 1991) and established the avian-like H1N1 lineage (avH1N1) that currently is the predominant H1N1 in the continent (Kyriakis et al., 2011). Other H1N1 strains of other origins have been isolated as well but did not establish in the swine population (reviewed by Kuntz-Simon and Madec, 2009)

With respect to the H3N2 subtype, this was detected firstly related to the 1968 human pandemic H3N2 (huH3N2) and is supposed to have been introduced in pigs from humans (Miwa et al., 1987). In fact, the huH3N2 isolates obtained during the following 16 years maintained a high genetic and antigenic similarity with seasonal human H3N2 isolates (Aymard et al., 1980; Ottis et al., 1982; Castrucci et al., 1994) suggesting a

Chapter 1

constant introduction of human viruses to the pig population more than an adaptation of the virus in pigs.

In 1984, a reassortment event between the avH1N1 and the huH3N2 resulted in the generation of a strain containing the internal genes of the avH1N1 and the glycoproteins of the huH3N2 (Castrucci et al., 1993). This virus, named “reassortant human like swine” H3N2 (rH3N2), was well adapted in pigs and spread in Europe. Thus in Spain, Italy, Denmark, Belgium, the Netherlands or Germany (Böttcher et al., 2007; Van Reeth et al., 2008; Simon-Grifé et al., 2011) the seroprevalence of H3N2 is moderate to high while the virus is absent in France, Great Britain and Ireland and cannot be detected by serology in the Czech Republic (Franck et al., 2007; Rosembergova et al., 2007;

Kyriakis et al., 2011). In Poland, the introduction of these new H3N2 is recent since in 2008 could not be detected (Van Reeth et al., 2008; Kowalczyk et al., 2010). The last SIV subtype emerging in Europe is the H1N2. The diversity observed within H1N2 isolates strains when compared with either the H1N1 or H3N2 European is high, and it is thought to be the subtype which has been implicated more reassortment events in Europe (reviewed by Kuntz-Simon and Madec, 2009). In fact, the first SIV H1N2 isolate obtained in Britain (France) in 1987 (avH1N2) was a reassortant strain which contained the HA of the avH1N1 and the other genes came from the huH3N2 (Gourreau et al., 1994). However, this SIV subtype was not established in the swine population until 1994. At that time emerged a new reassortant harboring HA related to the Russian flu H1N1 of 1977 (see section 1.4.2) and the rest of the gene segments comes from the rhsH3N2 virus (named “reassortant human-like swine” H1N2 (rH1N2) (Brown et al., 1995). The HA of those H1N2 does not present any cross reactivity with the avH1N1 HA and both viruses can be differentiated by serological assays (Brown et al., 1998)

Chapter 1

while other H1N2 of different origin have been detected as well. For example, H1N2 viruses containing the NA segment of the “seasonal-human” H3N2 have been found in Italy (Moreno et al., 2012), others carrying a HA related to the avH1N1 virus found in Denmark and France (Hjulsager et al., 2006; Kuntz-Simon and Madec, 2009) and recently one H1N2 isolated in Sweden and Italy harboring the NA from the rH3N2 than the rH1N2 (Bálint et al., 2009; Moreno et al., 2012).

1.4.3.3. SIV in Asia

Epidemiology of Asian influenza viruses of pigs present a number of particularities compared to the infection occurring in other regions. This is probably attributable to the existence of a very large pig population often raised under extensive systems and in contact with migratory birds. The first particularity of Asian SIV is the existence of two main lineages within the H1N1 subtype (csH1N1 and avH1N1) as well as a wide variety of different reassortants belonging to this subtype (Vijaykrishna et al., 2011;

Choi et al., 2012). The diversity of H3N2 strains in Asia is more complex than in Europe or North America. In fact, in Asia there are H3N2 reassortants between huH3N2, rH3N2 and the North American triple reassortant H3N2 (trH3N2; see section 1.4.3.4). As a matter of fact, H3N2 SIV isolated in China and Thailand have never been reported elsewhere (Chutinimitkul et al., 2008; Takemae et al., 2008; Yu et al., 2008).

To add complexity to this picture, since 2006 there are no evidences for the circulation of the H3N2 subtype in China (Vijaykrishna et al., 2011) although that H3N2 circulates in Korea and other Asian countries(Song et al., 2003; Lee et al., 2008). Taking into account that China is the first pig producer of the world, this is not a negligible fact

Chapter 1

In Asia H1N2 was firstly isolated in 1978 (Japan), earlier than in Europe and North America. In that case, the H1N2 virus was a reassortant between the csH1N1 and seasonal human H3N2 viruses. That reassortant established a lineage in Japan which has been circulating in that country since then (Yoneyama et al., 2010). In Korea a triple reassortant H1N2 related to the North American viruses has been circulating since 2002 (Choi et al., 2002; Pascua et al., 2008). In China, strains belonging to the North American triple reassortant H1N2 (trH1N2, see section 1.4.3.4), rH1N2, and reassortants of both strains have been described (Yu et al., 2009; Vijaykrishna et al., 2011).

1.4.3.4. SIV in North America

Influenza virus isolated in North America or elsewhere (1931) was an H1N1 strain of pigs closely related with the Spanish flu H1N1 of 1918 (Shope , 1931) (Figure 6). In fact, in 1919 a first description of a flu-like disease was observed affecting pigs in North America close in time to the spread Spanish Flu (Koen et al., 1919). The descendants of that H1N1 strain established what is called the classical swine H1N1 lineage that still circulates nowadays presenting different antigenic variants (reviewed by Olsen et al., 2002).

The H3N2 subtype was first detected detected in pigs of North America in last years of the 1970 decade, having a low clinical impact (Hinshaw et al., 1978). Introduction of H3N2 viruses probably took place as indicated by serologica evidences in Canada (Bikour et al., 1995). Those Canadian H3N2 strains were close to the seasonal human H3N2 strains isolated in 1975 (Bikour et al., 1994; Bikour et al., 1995). In 1997, a

Chapter 1

reassortment between the csH1N1 and the seasonal human H3N2 resulted in a new H3N2 strain which changed drastically the presence and the clinical impact of the H3N2 viruses in North American farms because of its higher virulence (Karasin et al, 2000a).

From that moment, the new H3N2 strain suffered reassortments which resulted in the introduction of internal gene segments (all but HA and NA) from avian and human influenza viruses circulating in North America in that time (Zhou et al., 1999; Karasin et al., 2000a). Of these, a triple pig-avian-human reassortant H3N2 (trH3N2) was established in pigs (Van Reeth et al., 2012).

The H1N2 subtype was firstly isolated in 1999 in North America. The original H1N2 SIV of North America harbored HA from the csH1N1 and the rest of the segments were of the trH3N2 virus. This virus has been described in several states of the USA and also in Canada and it is known as the triple reassortant H1N2 virus (trH1N2) (Karasin et al., 2000b). Additionally to these triple reassortant H3N2 and H1N2, since 2000, viruses presenting the HA and NA from csH1N1 viruses have been detected (Yassine et al., 2009) and are named as trH1N1 viruses.

Additionally, different authors have proposed the existence of different clusters based in the HA phylogeny and antigenic recognition of the North American SIV. There are 4 main differentiated clusters for the H1 strains (α, β, δ, γ) and for the H3 strains (I-IV) (Vincent et al., 2009; Lorusso et al., 2011; Kumar et al., 2011). However, whether this classification have impact in the epidemiology, pathogenesis and cross protection have not been deeply studied.

Figure 5. Reassortment events involved in the evolution of European SIV and year of detection of the reassortants. Black ovals indicate the reasssortment point. Gene abbreviations above and below the oval indicate the contribution of each virus involved in the reassortment event.

Figure 6. Reassortment events involved in the evolution of North American SIV and year of detection of the reassortants. Black ovals indicate the reasssortment point. Gene abbreviations above and below the oval indicate the contribution of each virus involved in the reassortment event.

Chapter 1

27 1.4.3.5. SIV in Latin America

Little is known about the status of SIV infection in Latin America and most of the data account for serological analysis using the haemagglutination inhibition test (HAI). This is case for Chile (Vicente et al., 1979) Brazil (Mancini et al., 2006 ; Rajao et al., 2011), and Venezuela (Boulanger et al., 2004). These works indicated that H1N1 and H3N2 swine strains related with the North American lineages are circulating in Latin America.

More recently, Cappuccio and coworkers (2011) described an H3N2 circulating in pigs in 2008. This virus was related with human seasonal H3N2 viruses circulating in 2003.

However, the number of SIV sequences available from Latin America is scarce and this makes difficult to draw adequate conclusions.

1.4.3.6. Other influenza subtypes in pigs

Infections by subtypes other than H1N1, H3N2 and H1N2 have been also documented sporadically in pigs. For example H1N7, H3N1, H3N3, H4N6, H4N8, H6N6, H7N2, H7N7 viruses have been reported in pigs either by isolation, PCR or by serology (Brown et al., 1994; Karasin et al., 2000c; Karasin et al., 2004; Lekcharoensuk et al., 2006; Moreno et al., 2009; de Jong et al., 2009; Kwon et al., 2011; Zhang et al., 2011;

Su et al., 2012). In none of those cases the viruses had any particular virulence. Other subtypes such H2N3, H9N2 and H5N1 deserve a more detailed comment.

The H2N3 subtype was isolated from two swine herds in Missouri (USA) (Ma et al., 2007). Under experimental conditions the virus was able to infect pigs, was transmissible among pigs and ferrets without prior adaptation and produced an overt

Dans le document Insights in the molecular epidemiology and antigenic characterization of (Page 38-58)